© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 6125
    Prev
    Next
    INDEX
    Explanations

    instances of intense emotional reactions or strong opinions

    oai_token-act-pair · gpt-3.5-turbo

    emotional expressions and reactions in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    upt
    -0.74
    adi
    -0.72
     carbohyd
    -0.72
    inse
    -0.69
    nton
    -0.69
    ogue
    -0.68
    è¦ļéĨĴ
    -0.67
    enium
    -0.66
    gins
    -0.65
    uler
    -0.64
    POSITIVE LOGITS
     Firstly
    1.18
     Why
    1.11
     Surely
    1.08
     Perhaps
    1.06
     Especially
    1.04
     Suppose
    1.03
     Because
    1.02
     Maybe
    1.00
     Something
    1.00
     Certainly
    0.99
    Activations Density 0.449%

    No Known Activations